This repository has been archived on 2023-10-09.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/nodes/COM_KeyingNode.cpp

312 lines
12 KiB
C++
Raw Normal View History

Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
/*
* 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.
*
* Copyright 2012, Blender Foundation.
Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
*/
#include "COM_KeyingNode.h"
#include "COM_ExecutionSystem.h"
#include "COM_KeyingOperation.h"
#include "COM_KeyingBlurOperation.h"
2012-05-29 14:55:01 +00:00
#include "COM_KeyingDespillOperation.h"
#include "COM_KeyingClipOperation.h"
Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
#include "COM_MathBaseOperation.h"
#include "COM_ConvertOperation.h"
Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
#include "COM_SetValueOperation.h"
#include "COM_DilateErodeOperation.h"
#include "COM_SetAlphaOperation.h"
#include "COM_GaussianAlphaXBlurOperation.h"
#include "COM_GaussianAlphaYBlurOperation.h"
2012-06-15 17:57:39 +00:00
KeyingNode::KeyingNode(bNode *editorNode) : Node(editorNode)
Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
{
2012-06-15 17:57:39 +00:00
/* pass */
Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
}
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
NodeOperationOutput *KeyingNode::setupPreBlur(NodeConverter &converter, NodeInput *inputImage, int size) const
Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
{
ConvertRGBToYCCOperation *convertRGBToYCCOperation = new ConvertRGBToYCCOperation();
convertRGBToYCCOperation->setMode(BLI_YCC_ITU_BT709);
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.addOperation(convertRGBToYCCOperation);
2018-06-17 17:05:29 +02:00
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.mapInputSocket(inputImage, convertRGBToYCCOperation->getInputSocket(0));
2018-06-17 17:05:29 +02:00
Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
CombineChannelsOperation *combineOperation = new CombineChannelsOperation();
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.addOperation(combineOperation);
Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
for (int channel = 0; channel < 4; channel++) {
SeparateChannelOperation *separateOperation = new SeparateChannelOperation();
separateOperation->setChannel(channel);
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.addOperation(separateOperation);
2018-06-17 17:05:29 +02:00
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.addLink(convertRGBToYCCOperation->getOutputSocket(0), separateOperation->getInputSocket(0));
2018-06-17 17:05:29 +02:00
Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
if (channel == 0 || channel == 3) {
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.addLink(separateOperation->getOutputSocket(0), combineOperation->getInputSocket(channel));
Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
}
else {
KeyingBlurOperation *blurXOperation = new KeyingBlurOperation();
blurXOperation->setSize(size);
blurXOperation->setAxis(KeyingBlurOperation::BLUR_AXIS_X);
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.addOperation(blurXOperation);
2018-06-17 17:05:29 +02:00
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
KeyingBlurOperation *blurYOperation = new KeyingBlurOperation();
blurYOperation->setSize(size);
blurYOperation->setAxis(KeyingBlurOperation::BLUR_AXIS_Y);
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.addOperation(blurYOperation);
2018-06-17 17:05:29 +02:00
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.addLink(separateOperation->getOutputSocket(), blurXOperation->getInputSocket(0));
converter.addLink(blurXOperation->getOutputSocket(), blurYOperation->getInputSocket(0));
converter.addLink(blurYOperation->getOutputSocket(0), combineOperation->getInputSocket(channel));
Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
}
}
2018-06-17 17:05:29 +02:00
Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
ConvertYCCToRGBOperation *convertYCCToRGBOperation = new ConvertYCCToRGBOperation();
convertYCCToRGBOperation->setMode(BLI_YCC_ITU_BT709);
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.addOperation(convertYCCToRGBOperation);
2018-06-17 17:05:29 +02:00
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.addLink(combineOperation->getOutputSocket(0), convertYCCToRGBOperation->getInputSocket(0));
2018-06-17 17:05:29 +02:00
Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
return convertYCCToRGBOperation->getOutputSocket(0);
}
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
NodeOperationOutput *KeyingNode::setupPostBlur(NodeConverter &converter, NodeOperationOutput *postBlurInput, int size) const
Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
{
KeyingBlurOperation *blurXOperation = new KeyingBlurOperation();
blurXOperation->setSize(size);
blurXOperation->setAxis(KeyingBlurOperation::BLUR_AXIS_X);
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.addOperation(blurXOperation);
2018-06-17 17:05:29 +02:00
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
KeyingBlurOperation *blurYOperation = new KeyingBlurOperation();
blurYOperation->setSize(size);
blurYOperation->setAxis(KeyingBlurOperation::BLUR_AXIS_Y);
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.addOperation(blurYOperation);
2018-06-17 17:05:29 +02:00
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.addLink(postBlurInput, blurXOperation->getInputSocket(0));
converter.addLink(blurXOperation->getOutputSocket(), blurYOperation->getInputSocket(0));
2018-06-17 17:05:29 +02:00
return blurYOperation->getOutputSocket();
Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
}
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
NodeOperationOutput *KeyingNode::setupDilateErode(NodeConverter &converter, NodeOperationOutput *dilateErodeInput, int distance) const
Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
{
DilateDistanceOperation *dilateErodeOperation;
Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
if (distance > 0) {
dilateErodeOperation = new DilateDistanceOperation();
dilateErodeOperation->setDistance(distance);
Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
}
else {
dilateErodeOperation = new ErodeDistanceOperation();
dilateErodeOperation->setDistance(-distance);
Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
}
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.addOperation(dilateErodeOperation);
2018-06-17 17:05:29 +02:00
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.addLink(dilateErodeInput, dilateErodeOperation->getInputSocket(0));
2018-06-17 17:05:29 +02:00
Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
return dilateErodeOperation->getOutputSocket(0);
}
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
NodeOperationOutput *KeyingNode::setupFeather(NodeConverter &converter, const CompositorContext &context,
NodeOperationOutput *featherInput, int falloff, int distance) const
{
/* this uses a modified gaussian blur function otherwise its far too slow */
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
CompositorQuality quality = context.getQuality();
/* initialize node data */
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
NodeBlurData data;
memset(&data, 0, sizeof(NodeBlurData));
data.filtertype = R_FILTER_GAUSS;
if (distance > 0) {
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
data.sizex = data.sizey = distance;
}
else {
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
data.sizex = data.sizey = -distance;
}
GaussianAlphaXBlurOperation *operationx = new GaussianAlphaXBlurOperation();
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
operationx->setData(&data);
operationx->setQuality(quality);
operationx->setSize(1.0f);
operationx->setSubtract(distance < 0);
operationx->setFalloff(falloff);
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.addOperation(operationx);
2018-06-17 17:05:29 +02:00
GaussianAlphaYBlurOperation *operationy = new GaussianAlphaYBlurOperation();
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
operationy->setData(&data);
operationy->setQuality(quality);
operationy->setSize(1.0f);
operationy->setSubtract(distance < 0);
operationy->setFalloff(falloff);
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.addOperation(operationy);
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.addLink(featherInput, operationx->getInputSocket(0));
converter.addLink(operationx->getOutputSocket(), operationy->getInputSocket(0));
return operationy->getOutputSocket();
}
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
NodeOperationOutput *KeyingNode::setupDespill(NodeConverter &converter, NodeOperationOutput *despillInput, NodeInput *inputScreen,
float factor, float colorBalance) const
Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
{
2012-05-29 14:55:01 +00:00
KeyingDespillOperation *despillOperation = new KeyingDespillOperation();
despillOperation->setDespillFactor(factor);
despillOperation->setColorBalance(colorBalance);
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.addOperation(despillOperation);
2018-06-17 17:05:29 +02:00
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.addLink(despillInput, despillOperation->getInputSocket(0));
converter.mapInputSocket(inputScreen, despillOperation->getInputSocket(1));
2018-06-17 17:05:29 +02:00
2012-05-29 14:55:01 +00:00
return despillOperation->getOutputSocket(0);
Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
}
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
NodeOperationOutput *KeyingNode::setupClip(NodeConverter &converter, NodeOperationOutput *clipInput, int kernelRadius, float kernelTolerance,
float clipBlack, float clipWhite, bool edgeMatte) const
{
KeyingClipOperation *clipOperation = new KeyingClipOperation();
clipOperation->setKernelRadius(kernelRadius);
clipOperation->setKernelTolerance(kernelTolerance);
clipOperation->setClipBlack(clipBlack);
clipOperation->setClipWhite(clipWhite);
clipOperation->setIsEdgeMatte(edgeMatte);
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.addOperation(clipOperation);
2018-06-17 17:05:29 +02:00
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.addLink(clipInput, clipOperation->getInputSocket(0));
2018-06-17 17:05:29 +02:00
return clipOperation->getOutputSocket(0);
}
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
void KeyingNode::convertToOperations(NodeConverter &converter, const CompositorContext &context) const
Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
{
bNode *editorNode = this->getbNode();
NodeKeyingData *keying_data = (NodeKeyingData *) editorNode->storage;
2018-06-17 17:05:29 +02:00
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
NodeInput *inputImage = this->getInputSocket(0);
NodeInput *inputScreen = this->getInputSocket(1);
NodeInput *inputGarbageMatte = this->getInputSocket(2);
NodeInput *inputCoreMatte = this->getInputSocket(3);
NodeOutput *outputImage = this->getOutputSocket(0);
NodeOutput *outputMatte = this->getOutputSocket(1);
NodeOutput *outputEdges = this->getOutputSocket(2);
NodeOperationOutput *postprocessedMatte = NULL, *postprocessedImage = NULL, *edgesMatte = NULL;
2018-06-17 17:05:29 +02:00
Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
/* keying operation */
KeyingOperation *keyingOperation = new KeyingOperation();
keyingOperation->setScreenBalance(keying_data->screen_balance);
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.addOperation(keyingOperation);
2018-06-17 17:05:29 +02:00
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.mapInputSocket(inputScreen, keyingOperation->getInputSocket(1));
2018-06-17 17:05:29 +02:00
Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
if (keying_data->blur_pre) {
/* chroma preblur operation for input of keying operation */
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
NodeOperationOutput *preBluredImage = setupPreBlur(converter, inputImage, keying_data->blur_pre);
converter.addLink(preBluredImage, keyingOperation->getInputSocket(0));
Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
}
else {
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.mapInputSocket(inputImage, keyingOperation->getInputSocket(0));
Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
}
2018-06-17 17:05:29 +02:00
postprocessedMatte = keyingOperation->getOutputSocket();
2018-06-17 17:05:29 +02:00
/* black / white clipping */
2012-06-15 17:57:39 +00:00
if (keying_data->clip_black > 0.0f || keying_data->clip_white < 1.0f) {
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
postprocessedMatte = setupClip(converter, postprocessedMatte,
keying_data->edge_kernel_radius, keying_data->edge_kernel_tolerance,
keying_data->clip_black, keying_data->clip_white, false);
}
2018-06-17 17:05:29 +02:00
/* output edge matte */
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
edgesMatte = setupClip(converter, postprocessedMatte,
keying_data->edge_kernel_radius, keying_data->edge_kernel_tolerance,
keying_data->clip_black, keying_data->clip_white, true);
2018-06-17 17:05:29 +02:00
/* apply garbage matte */
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
if (inputGarbageMatte->isLinked()) {
SetValueOperation *valueOperation = new SetValueOperation();
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
valueOperation->setValue(1.0f);
converter.addOperation(valueOperation);
2018-06-17 17:05:29 +02:00
MathSubtractOperation *subtractOperation = new MathSubtractOperation();
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.addOperation(subtractOperation);
2018-06-17 17:05:29 +02:00
MathMinimumOperation *minOperation = new MathMinimumOperation();
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.addOperation(minOperation);
2018-06-17 17:05:29 +02:00
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.addLink(valueOperation->getOutputSocket(), subtractOperation->getInputSocket(0));
converter.mapInputSocket(inputGarbageMatte, subtractOperation->getInputSocket(1));
2018-06-17 17:05:29 +02:00
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.addLink(subtractOperation->getOutputSocket(), minOperation->getInputSocket(0));
converter.addLink(postprocessedMatte, minOperation->getInputSocket(1));
2018-06-17 17:05:29 +02:00
postprocessedMatte = minOperation->getOutputSocket();
}
2018-06-17 17:05:29 +02:00
/* apply core matte */
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
if (inputCoreMatte->isLinked()) {
MathMaximumOperation *maxOperation = new MathMaximumOperation();
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.addOperation(maxOperation);
2018-06-17 17:05:29 +02:00
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.mapInputSocket(inputCoreMatte, maxOperation->getInputSocket(0));
converter.addLink(postprocessedMatte, maxOperation->getInputSocket(1));
2018-06-17 17:05:29 +02:00
postprocessedMatte = maxOperation->getOutputSocket();
}
2018-06-17 17:05:29 +02:00
Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
/* apply blur on matte if needed */
if (keying_data->blur_post)
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
postprocessedMatte = setupPostBlur(converter, postprocessedMatte, keying_data->blur_post);
Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
/* matte dilate/erode */
if (keying_data->dilate_distance != 0) {
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
postprocessedMatte = setupDilateErode(converter, postprocessedMatte, keying_data->dilate_distance);
Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
}
/* matte feather */
if (keying_data->feather_distance != 0) {
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
postprocessedMatte = setupFeather(converter, context, postprocessedMatte, keying_data->feather_falloff,
keying_data->feather_distance);
}
Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
/* set alpha channel to output image */
SetAlphaOperation *alphaOperation = new SetAlphaOperation();
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.addOperation(alphaOperation);
2018-06-17 17:05:29 +02:00
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.mapInputSocket(inputImage, alphaOperation->getInputSocket(0));
converter.addLink(postprocessedMatte, alphaOperation->getInputSocket(1));
Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
postprocessedImage = alphaOperation->getOutputSocket();
2012-05-29 14:55:01 +00:00
/* despill output image */
if (keying_data->despill_factor > 0.0f) {
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
postprocessedImage = setupDespill(converter, postprocessedImage,
inputScreen,
keying_data->despill_factor,
keying_data->despill_balance);
Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
}
/* connect result to output sockets */
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.mapOutputSocket(outputImage, postprocessedImage);
converter.mapOutputSocket(outputMatte, postprocessedMatte);
if (edgesMatte)
Structural cleanup and improvements for the compositor. Many parts of the compositor are unnecessarily complicated. This patch aims at reducing the complexity of writing nodes and making the code more transparent. == Separating Nodes and Operations == Currently these are both mixed in the same graph, even though they have very different purposes and are used at distinct stages in the compositing process. The patch introduces dedicated graph classes for nodes and for operations. This removes the need for a lot of special case checks (isOperation etc.) and explicit type casts. It simplifies the code since it becomes clear at every stage what type of node we are dealing with. The compiler can use static typing to avoid common bugs from mixing up these types and fewer runtime sanity checks are needed. == Simplified Node Conversion == Converting nodes to operations was previously based on "relinking", i.e. nodes would start with by mirroring links in the Blender DNA node trees, then add operations and redirect these links to them. This was very hard to follow in many cases and required a lot of attention to avoid invalid states. Now there is a helper class called the NodeConverter, which is passed to nodes and implements a much simpler API for this process. Nodes can add operations and explicit connections as before, but defining "external" links to the inputs/outputs of the original node now uses mapping instead of directly modifying link data. Input data (node graph) and result (operations graph) are cleanly separated. == Removed Redundant Data Structures == A few redundant data structures have been removed, notably the SocketConnection. These are only needed temporarily during graph construction. For executing the compositor operations it is perfectly sufficient to store only the direct input link pointers. A common pointer indirection is avoided this way (which might also give a little performance improvement). == Avoid virtual recursive functions == Recursive virtual functions are evil. They are very hard to follow during debugging. At least in the parts this patch is concerned with these functions have been replaced by a non-virtual recursive core function (which might then call virtual non-recursive functions if needed). See for example NodeOperationBuilder::group_operations.
2014-04-15 16:06:12 +02:00
converter.mapOutputSocket(outputEdges, edgesMatte);
Initial commit of new keying nodes First node is called Keying Screen (Add -> Matte -> Keying Screen) and it's aimed to resolve issues with gradients on green screens by producing image with gradient which is later used as an input for screen color in keying nodes. This node gets motion tracks from given movie clip and trackign object and uses them to define color and position of points of gradient: for position marker's position on current frame is sued, for color average color of pattern area is used. Gradient is calculating in the following way: - On first step voronoi diagram is creating for given tracks. - On second step triangulation of this diagram happens by connecting sites to edges which defines area this site belongs to. - On third step gradient filling of this triangles happens. One of triangle vertices is colored with average track color, two rest vertoces are colored with average color between two neighbor sites. Current pixel's color in triangle is calculating as linear combination of vertices colors and barycentric coordinates of this pixel. This node is implemented for both tile and legacy compositor systems. Second node is basically a combination of several existing nodes to make keying more straighforward and reduce spagetti mess in the compositor, but it also ships some fresh approaches calculating matte which seems to be working better for not actually green screens. This node supports: - Chroma preblur - Dispilling - Clip white/black - Dilate/Erode - Matte post blur This node doesn't support chroma pre-blur for legacy compositor (yet). There're still lots of stuff to be improved here, but this nodes night already be used i think. Some details might be found on this wiki page: http://wiki.blender.org/index.php/User:Nazg-gul/Keying This patch also contains some currently unused code from color math module, but it was used for tests and might be used for tests in the future. Think it's ok to have it in branch at least.
2012-05-29 14:00:47 +00:00
}