geometrize::Model Class Reference

The Model class is the model for the core optimization/fitting algorithm. More...

#include <model.h>

Classes
class	ModelImpl

Public Member Functions
	Model (const geometrize::Bitmap &target)
	Model Creates a model that will aim to replicate the target bitmap with shapes. More...
	Model (const geometrize::Bitmap &target, const geometrize::Bitmap &initial)
	Model Creates a model that will optimize for the given target bitmap, starting from the given initial bitmap. The target bitmap and initial bitmap must be the same size (width and height). More...
	~Model ()
Model &	operator= (const Model &)=delete
	Model (const Model &)=delete
void	reset (geometrize::rgba backgroundColor)
	reset Resets the model back to the state it was in when it was created. More...
std::int32_t	getWidth () const
	getWidth Gets the width of the target bitmap. More...
std::int32_t	getHeight () const
	getHeight Gets the height of the target bitmap. More...
std::vector< geometrize::ShapeResult >	step (const std::function< std::shared_ptr< geometrize::Shape >(void)> &shapeCreator, std::uint8_t alpha, std::uint32_t shapeCount, std::uint32_t maxShapeMutations, std::uint32_t maxThreads, const geometrize::core::EnergyFunction &energyFunction=nullptr, const geometrize::ShapeAcceptancePreconditionFunction &addShapePrecondition=nullptr)
	step Steps the primitive optimization/fitting algorithm. More...
geometrize::ShapeResult	drawShape (std::shared_ptr< geometrize::Shape > shape, geometrize::rgba color)
	drawShape Draws a shape on the model. Typically used when to manually add a shape to the image (e.g. when setting an initial background). NOTE this unconditionally draws the shape, even if it increases the difference between the source and target image. More...
geometrize::Bitmap &	getCurrent ()
	getCurrent Gets the current bitmap. More...
geometrize::Bitmap &	getTarget ()
	getTarget Gets the target bitmap. More...
const geometrize::Bitmap &	getCurrent () const
	getCurrent Gets the current bitmap, const-edition. More...
const geometrize::Bitmap &	getTarget () const
	getTarget Gets the target bitmap, const-edition. More...
void	setSeed (std::uint32_t seed)
	setSeed Sets the seed that the random number generators of this model use. Note that the model also uses an internal seed offset which is incremented when the model is stepped. More...

Private Attributes
std::unique_ptr< Model::ModelImpl >	d

Detailed Description

The Model class is the model for the core optimization/fitting algorithm.

Author

Sam Twidale (https://samcodes.co.uk/)

Constructor & Destructor Documentation

Model() [1/3]

geometrize::Model::Model

(

const geometrize::Bitmap &

target

)

Model Creates a model that will aim to replicate the target bitmap with shapes.

Parameters

target

The target bitmap to replicate with shapes.

                                           : d{std::unique_ptr<Model::ModelImpl>(new Model::ModelImpl(target))}
{}

Model() [2/3]

geometrize::Model::Model	(	const geometrize::Bitmap &	target,
		const geometrize::Bitmap &	initial
	)

Model Creates a model that will optimize for the given target bitmap, starting from the given initial bitmap. The target bitmap and initial bitmap must be the same size (width and height).

Parameters

target	The target bitmap to replicate with shapes.
initial	The starting bitmap.

                                                                            : d{std::unique_ptr<Model::ModelImpl>(new Model::ModelImpl(target, initial))}
{}

~Model()

geometrize::Model::~Model

(

)

{}

Model() [3/3]

geometrize::Model::Model ( const Model &  )

delete

Member Function Documentation

drawShape()

geometrize::ShapeResult geometrize::Model::drawShape	(	std::shared_ptr< geometrize::Shape >	shape,
		geometrize::rgba	color
	)

drawShape Draws a shape on the model. Typically used when to manually add a shape to the image (e.g. when setting an initial background). NOTE this unconditionally draws the shape, even if it increases the difference between the source and target image.

Parameters

shape	The shape to draw.
color	The color (including alpha) of the shape.

Returns

Data about the shape drawn on the model.

{
    return d->drawShape(shape, color);
}

getCurrent() [1/2]

geometrize::Bitmap & geometrize::Model::getCurrent

(

)

getCurrent Gets the current bitmap.

Returns

The current bitmap.

{
    return d->getCurrent();
}

Here is the caller graph for this function:

getCurrent() [2/2]

const geometrize::Bitmap & geometrize::Model::getCurrent

(

)

const

getCurrent Gets the current bitmap, const-edition.

Returns

The current bitmap.

{
    return d->getCurrent();
}

getHeight()

std::int32_t geometrize::Model::getHeight

(

)

const

getHeight Gets the height of the target bitmap.

Returns

The height of the target bitmap.

{
    return d->getHeight();
}

getTarget() [1/2]

geometrize::Bitmap & geometrize::Model::getTarget

(

)

getTarget Gets the target bitmap.

Returns

The target bitmap.

{
    return d->getTarget();
}

Here is the caller graph for this function:

getTarget() [2/2]

const geometrize::Bitmap & geometrize::Model::getTarget

(

)

const

getTarget Gets the target bitmap, const-edition.

Returns

The target bitmap.

{
    return d->getTarget();
}

getWidth()

std::int32_t geometrize::Model::getWidth

(

)

const

getWidth Gets the width of the target bitmap.

Returns

The width of the target bitmap.

{
    return d->getWidth();
}

operator=()

Model & geometrize::Model::operator= ( const Model &  )

delete

reset()

void geometrize::Model::reset

(

geometrize::rgba

backgroundColor

)

reset Resets the model back to the state it was in when it was created.

Parameters

backgroundColor

The starting background color to use.

{
    d->reset(backgroundColor);
}

setSeed()

void geometrize::Model::setSeed

(

std::uint32_t

seed

)

setSeed Sets the seed that the random number generators of this model use. Note that the model also uses an internal seed offset which is incremented when the model is stepped.

Parameters

seed	The random number generator seed.

{
    d->setSeed(seed);
}

Here is the caller graph for this function:

step()

std::vector< geometrize::ShapeResult > geometrize::Model::step	(	const std::function< std::shared_ptr< geometrize::Shape >(void)> &	shapeCreator,
		std::uint8_t	alpha,
		std::uint32_t	shapeCount,
		std::uint32_t	maxShapeMutations,
		std::uint32_t	maxThreads,
		const geometrize::core::EnergyFunction &	energyFunction = nullptr,
		const geometrize::ShapeAcceptancePreconditionFunction &	addShapePrecondition = nullptr
	)

step Steps the primitive optimization/fitting algorithm.

Parameters

shapeCreator	A function that will produce the shapes.
alpha	The alpha of the shape.
shapeCount	The number of random shapes to generate (only 1 is chosen in the end).
maxShapeMutations	The maximum number of times to mutate each random shape.
maxThreads	The maximum number of threads to use during this step.
energyFunction	An optional function to calculate the energy (if unspecified a default implementation is used).
addShapePrecondition	An optional function to determine whether to accept a shape (if unspecified a default implementation is used).

Returns

A vector containing data about the shapes added to the model in this step. This may be empty if no shape that improved the image could be found.

{
    return d->step(shapeCreator, alpha, shapeCount, maxShapeMutations, maxThreads, energyFunction, addShapePrecondition);
}

Here is the caller graph for this function:

Member Data Documentation

d

std::unique_ptr<Model::ModelImpl> geometrize::Model::d

private

---

The documentation for this class was generated from the following files:

• /home/appveyor/projects/geometrize-lib-docs/geometrize-lib/geometrize/geometrize/model.h
• /home/appveyor/projects/geometrize-lib-docs/geometrize-lib/geometrize/geometrize/model.cpp